In this application we describe our plans to create a Molecular Target Discovery and Development Center (MTDDC) that will act as downstream component of The Cancer Genome Atlas (TCGA) project. Our premise is that the complexity of cancer genome alterations leads directly to the heterogeneity of cancer behavior and outcome, and that to translate the wealth of cancer genome characterization into clinical utility requires the functional identification and validation of the underlying driver genes. Driver gene identification will lead to a deeper understanding of cancer genotypes, create an important new set of biomarkers and therapeutic targets, and when combined with genome-wide RNAi screens, lead to the identification of key genetic vulnerabilities that will serve as a new generation of therapeutic targets. Our planned center is a natural expansion of long- standing collaborative projects at Cold Spring Harbor Laboratory (CSHL) and combines several powerful methods that we have developed and will continue to build upon as outlined in this application. These methods include flexible mouse models based on the transplantation of genetically-manipulated progenitor cells into the appropriate tissues of recipient mice;novel bioinformatics that take complex cancer genome datasets and pinpoint candidate driver genes and considerably altered pathways;new RNAi technology to manipulate the expression of candidate target genes in vitro and in vivo;and genome-wide RNAi screens to find genetic vulnerabilities of cancer cells. The CSHL MTDDC will use these innovative tools to place the complex array of genomic alterations identified by cancer genome projects into biologic context. High-throughput screening in mouse models will be used to determine whether candidate genes are drivers or passengers. Additionally, through the identification of those driver genes that are required for tumor maintenance and by genome-wide RNAi screens to find the druggable vulnerabilities of major cancer genotypes, we will discover and validate a new generation of cancer drug targets. The resultant data, reagents, and newly validated biomarkers and targets will be openly shared among the TCGA network and broader cancer research communities, as we have done with RNAi Codex, CSHL's open-access portal/database for short-hairpin RNA (shRNA) gene-silencing constructs. PUBLIC HEALTH RELEVANCE: There are two critical goals for Molecular Target Discovery and Development Centers to make towards helping develop new treatments for cancer. The first goal must be to validate which of the gene mutations is a driver for cancer progression. These markers may lead to therapeutic targets, but this is unlikely for most genes. However the driver genes have great potential for diagnosis, prognosis and linking therapies to cancer subtypes. It is also necessary, however, to discover gene dependencies, the genes that are validated targets for cancers of specific genetic make up or tissue type. Such dependencies have the potential to become therapeutic targets that will selective target cancer cells versus normal cells.